Hello,
I tested the new leasqr and dfdp functions with the new 'bounds' feature and it
seems to work fine to me.
Thanks,
Bertrand
---
http://broessli.redbubble.com/
--- On Tue, 8/11/09, Søren Hauberg wrote:
From: Søren Hauberg
Subject: Re: [OctDev] package contribution
To: "Olaf Till"
C
Hello,
Just have a comment which is also a question concerning dfdp.m
The help states
"dp= fractional increment of p for numerical derivatives"
which is fine, but the problem comes that usually nobody knows
which step to use to calculate the derivatives.
Would it no be better to choose the s
Hi,
Thanks, I will make the changes. It will take some time tough...
Regarding the "bounds" in nmsmax I did it in a primitive way
that is enough for my needs:
[f(1)] = dirn*feval(fun,x,varargin{:});
--> [f(1), param] = dirn*feval(fun,x,varargin{:});
where the function fun renormalise the para
, roessli bertrand wrote:
> > > -I added a varargin variable to leasqr.m to allow pass additional
> > > parameters.
>
> "This should be unnecessary since one can use anonymous functions for
> passing additional parameters to the user function. "
>
> Can yo
09 at 11:25:57AM +0200, Søren Hauberg wrote:
> man, 03 08 2009 kl. 00:34 -0700, skrev roessli bertrand:
> > In the optim package:
> > ---
> > Mufit uses nmsmax, leasqr and samin.cc and I modified these functions
> > sligthly
> > to fi
"The changes you have made are they general purpose?"
I hope so, but somebody should check.
Bertrand
---
http://broessli.redbubble.com/
--- On Mon, 8/3/09, Søren Hauberg wrote:
From: Søren Hauberg
Subject: Re: [OctDev] package contribution
To: "roessli bertrand&q
eagrds,
Bertrand Roessli
---
http://broessli.redbubble.com/
--- On Sat, 8/1/09, Søren Hauberg wrote:
From: Søren Hauberg
Subject: Re: [OctDev] package contribution
To: "roessli bertrand"
Cc: octave-dev@lists.sourceforge.net
Date: Saturday, August 1, 2009, 5:26 PM
Hi,
fre, 31 0
Hello,
I have been trying to use bfgsmin, but the fit diverges for e.g. an exponential
fit:
---warning: __bfgsmin_obj: objective function could not be evaluated - setting
to DBL_MAX
error: feval: function `__isequal__' not found
I think the problem comes from
check_gradient.column(0) = g;
a
Hello,
I have been using the very useful
vrml-1.0.10-package from Octave-forge and
came across 2 small problems.
1.) the function vrlm_transfo calls the orthogonalize
function. I think this function was replaced by the
function orth in the linear-algebra package? Is it the case?
2.-) in t
